ignore-walk

Nested/recursive .gitignore/.npmignore parsing and filtering.

Walk a directory creating a list of entries, parsing any .ignore files met along the way to exclude files.

USAGE

const walk = require('ignore-walk')

// All options are optional, defaults provided.

// this function returns a promise, but you can also pass a cb
// if you like that approach better.
walk({
  path: '...', // root dir to start in. defaults to process.cwd()
  ignoreFiles: [ '.gitignore' ], // list of filenames. defaults to ['.ignore']
  includeEmpty: true|false, // true to include empty dirs, default false
  follow: true|false // true to follow symlink dirs, default false
}, callback)

// to walk synchronously, do it this way:
const result = walk.sync({ path: '/wow/such/filepath' })

If you want to get at the underlying classes, they're at walk.Walker and walk.WalkerSync.

OPTIONS

• path The path to start in. Defaults to process.cwd()

• ignoreFiles Filenames to treat as ignore files. The default is ['.ignore']. (This is where you'd put .gitignore or .npmignore or whatever.) If multiple ignore files are in a directory, then rules from each are applied in the order that the files are listed.

• includeEmpty Set to true to include empty directories, assuming they are not excluded by any of the ignore rules. If not set, then this follows the standard git behavior of not including directories that are empty.

  Note: this will cause an empty directory to be included if it would contain an included entry, even if it would have otherwise been excluded itself.

  For example, given the rules * (ignore everything) and !/a/b/c (re-include the entry at /a/b/c), the directory /a/b will be included if it is empty.

• follow Set to true to treat symbolically linked directories as directories, recursing into them. There is no handling for nested symlinks, so ELOOP errors can occur in some cases when using this option. Defaults to false.